1. Field of the Invention
The present invention relates to a technique for correcting optical characteristics in an image forming apparatus that employs an electrostatic recording method or an electrophotographic recording method.
2. Description of the Related Art
A conventional optical scanning apparatus is, for example, employed in an image forming apparatus of an electrophotographic type and configured to hold a constant level of laser light quantity during one scanning operation. To this end, a conventional control operation includes detecting an output of the laser within a predetermined light detection period (i.e., a beam detection (BD) period) being set during one scanning operation and holding the driving current for the laser at a constant level during one scanning operation. This is generally referred to as auto power control (APC) processing. One scanning operation is a single scanning operation of a laser beam in a longitudinal direction (i.e., an axial direction) of a photosensitive member (i.e., an image carrier).
However, when one scanning operation of the laser beam is performed on a photosensitive member (i.e., an image carrier), the density of a reproduced image varies depending on a laser position during the scanning operation. The unevenness of density is particular when images are compared between the center and the edge of the photosensitive member.
In general, a luminous flux incident on a polygonal mirror of an optical scanning apparatus has a Gauss distribution such that the light intensity can be maximized in the vicinity of an optical axis of a collecting optical system. Therefore, a light reflection and deflection region changes from the vicinity of the optical axis toward the edge according to an angle of field (i.e., a scanning angle capable of assuring an effective writing width relative to a photosensitive member).
The illuminance on a photosensitive member (i.e., a surface to be scanned) tends to decrease at the edge compared to the center. Namely, the illuminance decreases according to an increase in the image height (i.e., the position on a photosensitive member). The image height is “0” at the center of the photosensitive member and has a plus or minus value when the position moves toward the edge. This phenomenon is hereinafter referred to as “light quantity falloff at edges.”
In addition, due to an erroneous setup of a light source (such as a laser), the position where the intensity of an incident luminous flux is maximized on a deflection surface of a polygonal mirror may deviate from the center of an effective luminous flux width (relative to a main scanning direction of the deflection surface) toward an edge of the effective luminous flux width. In this case, in addition to light quantity falloff at edges, the illuminance on a scanned surface tends to increase or decrease when the position moves from one image height to another image height.
The illuminance along a scanning line on a photosensitive member (i.e., a surface to be scanned) may cause unevenness. Therefore, a formed image may have unevenness of density.
As discussed in Japanese Patent Application Laid-Open No. 2006-069118, in order to solve the above-described problem (i.e., to correct light quantity fall at edges), a conventional system divides one scanning period into a plurality of blocks and stores an amount of correction of light quantity falloff at edges (hereinafter referred to as “profile data”) for each block. During one scanning operation, the system reads profile data of a target block and profile data of a neighboring block, and controls a driving current value of the laser using linearly interpolated profile data obtained from the read data to correct light quantity falloff at edges.
FIG. 12 illustrates a conventional correction circuit for correcting light quantity falloff at edges. The correction circuit includes a digital/analog (D/A) converter 602 and a low-pass filter 604.
The D/A converter 602 receives a sampled-and-held voltage value VSH (i.e., a voltage value corresponding to a maximum light quantity) as a reference voltage to perform APC processing in a light detection period (i.e., the BD period) during one scanning operation. The D/A converter 602 also receives light quantity correction data 203 from a light quantity correction unit (not illustrated) and a clock signal CLK. The D/A converter 602 performs digital/analog conversion processing on the voltage value VSH based on the light quantity correction data 203 and the clock signal CLK.
The low-pass filter 604 is, for example, composed of a capacitor and a resistor. The low-pass filter 604 can filter an analog signal S605 received from the D/A converter 602. The filter 604 outputs an analog signal VCOM (i.e., a filtered signal) that can be used to control the current of a pulse current source of a laser drive control unit (not illustrated). Thus, the correction circuit can variably control the quantity of light during one scanning operation and can correct light quantity falloff at edges.
However, the following problem arises when the above-described conventional circuit corrects the driving current of a laser according to an amount of correction of light quantity falloff at edges of an optical system to solve the unevenness of density of an image.
More specifically, even if the rate of light quantity falloff at edges is 10% or 20%, i.e., even if the light quantity deteriorates from 100% to 90% or 80% during one scanning operation, a dynamic range for the D/A conversion is set to the light quantity of 100% to 0%.
For example, the above-described dynamic range setting is required to realize the correction of light quantity falloff at edges for a low light quantity of approximately 50%. As 1LSB (i.e., a minimum resolution of D/A) is a value determined in relation to the 100% light quantity, the resolution deteriorates according to a decrease in the light quantity. To solve this problem, the D/A of a higher resolution is required. Furthermore, re-calculation of correction data is required if there is any change in the light quantity to be used. Thus, the control becomes complicated.
The above-described problem is described in more detail with reference to FIGS. 11A and 11B. FIG. 11A illustrates a light quantity distribution on a drum surface including a BD image height (i.e., an image height at a BD position) at a light quantity level of P0. In FIG. 11A. P0BD represents a light quantity at the BD image height, PX0 represents a light quantity at an image height X0, . . . , and PXN represents a light quantity at an image height XN. Furthermore, (P0-PX0)/PX0, . . . , and (P0-PXN)/PXN represent correction amounts Y0, . . . , and YN at respective image heights X0 . . . XN.
Namely, each of the correction amounts Y0, . . . , and YN is a ratio of a difference between a target light quantity and an actual light quantity to the actual light quantity at each image height. The light quantity correction is performed by adding the correction amount to the laser driving current according to the image height, to have the target light quantity at each image height, as indicated by a corrected profile P′0 in FIG. 11A. In this case, if a D/A converter has a resolution of 8 bits, the minimum resolution 1LSB becomes P0/255. If the rate of light quantity falloff at edges at the light quantity level P0 is 20%, the correction data is somewhere in a range of 0 to 51(=255×20%) at each image height.
FIG. 11B illustrates a comparative light quantity distribution on a drum surface at a light quantity level of 0.5P0 (i.e., a half of the light quantity P0). In this case, the rate of light quantity falloff at edges is reduced to a half value. As illustrated in FIG. 11B, correction amounts Y′0, . . . , and Y′N corresponding to the light quantity level of 0.5P0 are (0.5P0-0.5PX0)/PX0=0.5(P0-PX0)/PX0=0.5Y0, . . . , and (0.5P0-0.5PXN)/PXN=0.5 (P0-PXN)/PXN=0.5YN at respective image heights.
Namely, if the light quantity in the acquisition of profile data is different from the actually used light quantity, the same profile data cannot be used. For example, if the rate of light quantity falloff at edges is 20% and an 8-bit D/A converter is used at the light quantity level of 0.5P0, the correction data is somewhere in a range of 0 to 26 (=255×20% x50%) at each image height. The resolution deteriorates largely compared to the correction range from 0 to 51, even if the correction rate for the light quantity is the same (20%). The resolution further deteriorates according to a decrease in the light quantity.